Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
In nearly every profession, past performance is indicative of future results. When picking a doctor, contractor, professor, attorney or professional athlete, naturally the key criterion to consider is their historical track record. The expertise of skilled professionals should be evidenced with consistent outperformance versus peers. Furthermore, the logic follows that anyone can suffer through a short-term setback for a variety of reasons, but the elite performers should not be expected to struggle for an extended period.
The same reasoning unfortunately does not apply to selecting investment managers. In fact, the counterintuitive nature of successful investing often requires acting against entrenched emotional biases that are shaped by the normal life experiences described above. We will describe why the conventional, instinctive approach to making investment manager choices generally leads to poor outcomes and how investors can be fooled by manager returns.
The traditional approach
If you were to ask a professional consultant or investor how they select investment managers, they would likely walk you through a well-rounded approach that combines “hard” data (strong historical performance) with “soft” analysis (qualitative evidence of superior skill). In practice, however, the manager’s returns tend to dominate the hire/fire decisions while less objective qualitative inputs often yield to the appeal of recent, precise data points. These are often reinforced by managers, most of whom are self-marketing experts, who ably weave compelling (and usually unprovable) stories to rationalize their recent success and why it should continue.
While all performance is usually considered, the convention is to focus on returns during the previous five years. Investors tend to assume that managers who have outperformed over a period as long as five years must be highly skilled and vice versa. It can be emotionally validating to hire a manager whose strategy “has been working” in the current environment. This further reinforces a short-term hard data approach to manager research. As just one example, Morningstar’s widely followed star rating also heavily weights historical five-year performance.1
The emphasis on short-term performance is heightened after the manager is hired and decisions are made about whether to retain or terminate the strategy. The return comparison to a relevant benchmark and peer group rankings typically dominate the evaluations of the manager’s success. This is in part due to a behavioral bias, to attach more value to information that is simple, immediate and precise. Whereas qualitative inputs can be ambiguous and more difficult to evaluate. When monitoring results this closely, five years may feel like an eternity. Very few investors exercise sufficient patience to hold on to an underperforming manager for several years, let alone five.
Consequently, the traditional approach typically results in hiring recent outperformers and firing underperformers. You may have personally experienced the frustrating phenomenon of adding a historically outperforming manager who suddenly underperforms just after being hired. By selling low, investors cement the poor returns into their portfolio and then risk hiring another potentially underperforming manager the next round. This buy-high, sell-low discipline clearly doesn’t work well over time.
Investing is different
Extrapolating the recent past in the hiring and firing of investment managers produces poor results because of a few unique characteristics that differentiate this industry from most others. First, consistent outperformance is extremely rare. We conducted a study2 of the returns of thousands of equity managers using Morningstar. We split the performance into four equal five-year segments over the past 20 years and ranked all the managers by quintile3 within their appropriate size and style category. Not a single equity fund in the universe ranked in the top quintile over all four five-year periods, demonstrating the rarity of consistent outperformance!
Remarkably, even the top long-term performers can underperform for long stretches. Warren Buffett is widely considered to be one of the greatest investors of all time. Most investors would be surprised to learn that Berkshire Hathaway has underperformed the Russell 1000 Index over the past 15 years! His 20-year track record outpaces the index, but all of the outperformance came in the first five years. Likewise, the single best performing equity funds within each of the nine Morningstar size and style boxes (large-, mid-, small-cap and growth, value and blend) all underperformed their relevant indices for periods greater than five years (and on average about 10 years). Table 1 summarizes the longest stretch of years the single best performer trailed the appropriate benchmark.4

The second aspect of investment management that is different from other professions is that performance over five years is an unreliable indicator of future returns. Since most investors are highly influenced by trailing five-year returns in their hire and fire decisions, we examined whether the data supported such a strategy. We reviewed how the top performers five years ago did in the subsequent five-year period. We compared those results to the managers who were the worst performers five years ago. In order to make relevant comparisons among managers, we separated the analysis into the nine Morningstar equity fund segments. Counterintuitively, in seven out of the nine categories the bottom quintile strategies outperformed the managers in the top quintile. In other words, in most cases you were more likely to find a future top performer if you were fishing from the bottom of the pool rather than the top.5
I have identified three key reasons why the practice of investment management is different from just about every other profession:
1. Skill is not required to outperform – an unskilled person can’t pitch a perfect game, but beating the market has an even chance of success by sheer luck. Randomly split the S&P 500 in half and one of the halves is essentially guaranteed to outperform the market.
2. Correct calls require recognition by the consensus to be rewarded – while a skilled manager may be able to buy a stock they know to be worth $100 for $50, profiting from that insight requires the consensus, which sets the market price, to catch on to the true value. Whereas a surgeon’s skill is evidenced, in real time, by each procedure she does, extended underperformance even for the best managers can result because of the randomness and unreliability of consensus views. To quote Benjamin Graham, “In the short term the market is a voting machine, in the long run it is a weighing machine.”
3. Investment styles go in and out of favor – almost every manager is biased to perform well in certain market environments and poorly in others. For instance, a manager whose skillset benefits from volatile investment conditions may justifiably underperform during a strong five-year bull market. Thus, highly skilled managers can underperform for an extended period simply because of an unconducive environment.
Focus on longer-term returns and qualitative analysis
The key takeaway is that investors constantly seek to improve their portfolios by identifying and fixing trouble spots. The assumption is that poor returns over a period as long as five years is sufficient evidence of an inferior manager that needs to be replaced. In reality, a more holistic approach to the critical hire/fire decision is prudent.
Short-term (five year) returns are a less useful guide than most investors recognize. However, this data point does offer some insight if evaluated within the appropriate context. When reviewing returns over a single up or down cycle, qualitatively evaluate whether the performance makes sense given the environment. That is, if the manager’s style/strategy is expected to underperform during the type of environment that transpired, then factor in that recognition when reviewing performance. This requires a deep understanding of the manager’s value add proposition, portfolio holdings and investment process. Of course, historical performance considered over multiple market cycles (10+ years) represents a more reliable indicator as the short-term effects of any one particular environment are netted out over time.
Ultimately, a qualitative appreciation of the manager should drive the investment decision. The superiority of the people and the process, as well as the demonstrated investment edge should foster greater confidence in the manager’s ability to beat its peers over the long run. The best managers offer the highest probability of producing future outperformance regardless of how they have done recently. If anything, a recently underperforming high-quality manager may present an excellent buying opportunity. Resisting the temptation to reflexively hire or fire managers based on recent returns can be challenging to implement in practice, but hopefully a better understanding of the flaws of the conventional approach will help move you one step in that direction.
Alex Shahidi and Damien Bisserier are partners at Advanced Research Investment Solutions, LLC (ARIS), a Los Angeles-based investment consulting firm.
1 Morningstar uses “risk-adjusted” historical returns to account for the level of risk taken to achieve a certain level of return. Nevertheless, a high return will generally earn a better rating than a low return.
2 The results of the study are for information and educational purposes only. It is not intended as an official statement.
3 Data source: Morningstar from 12/31/97 to 12/31/17.
4 This analysis excluded the top performer if the assets under management for the fund fell below $250 million.
5 To determine which quintile outperformed from 2012-2017, we compared the percentage of the top quintile managers from 2007-2011 versus those in the bottom quintile during the same period. Some may argue that the percentage of the bottom quintilers that jumped to the top quintile may be overstated because of survivorship bias (i.e., some of the underperforming managers may have dropped out of the universe). However, this does not necessarily mean that the managers that dropped out would not have outperformed had they remained in the universe.
Read more articles by Alex Shahidi, Damien Bisserier